{"id":"7dd2819e-01cc-4c90-be87-afb838fe624b","arxiv_id":"hep-ph/0004004","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"In perturbative QCD factorization, penguin amplitudes in B to K pi decays come out roughly 50% larger than factorization-assumption values, and annihilation diagrams contribute strong phases near 90 degrees.","lead":"The authors compute B to K pi decay amplitudes with perturbative QCD factorization and find penguin contributions about 50% larger than in the standard factorization approximation, plus annihilation diagrams that generate strong phases near 90 degrees. The result matters because it changes how the CP-violating phase gamma is extracted from B factory data.","discovery_kind":"unclear","skeptic_critique":{"model":"selfhosted/deepseek-v4.1-flash","headline":"Eq. (16)'s 50% penguin enhancement is a ratio of the same PQCD integral with running versus frozen Wilson coefficients, so it measures the Eq. (9) scale choice rather than an independent dynamical effect, and no scale-variation band is quoted.","rationale":"The reader's weakest_assumption correctly identifies the Eq. (9) scale-setting prescription as the load-bearing premise, and I would sharpen it: Eq. (16) is not an independent dynamical calculation but a ratio of the same integral with running versus frozen Wilson coefficients, so it is a direct measure of the scale choice. The reader's other points are also fair: Table II carries no theoretical error bars and is quoted for one representative parameter set with the full scan explicitly deferred; the wave functions are fitted to B→Dπ, ππ and the pion form factor, so the Kπ agreement is a partial test. I would give the paper credit on two counts. First, the Kπ prediction is a genuine out-of-sample test, since the wave functions were fixed on Dπ/ππ and the pion form factor, not on Kπ. Second, the paper is honest about the deferred parameter study and about the ~20% strong-phase uncertainty. The perturbativity check (α_s(t)/π < 0.3) is somewhat loose — at t ~ 1.4 GeV, α_s/π ≈ 0.16, so the threshold is satisfied almost everywhere above ~0.5 GeV — but this is secondary to the scale concern. None of these points breaks the central claim; they justify keeping the verdict at CONDITIONAL rather than ACCEPT. I therefore agree with the reader's CONDITIONAL and do not move the verdict.","tokens_in":10742,"tokens_out":5170,"duration_ms":99647,"concrete_test":"Recompute Eq. (16) and Table II with the hard scale in Eq. (9) varied by a factor of two (t/2 and 2t), the standard scale-variation estimate. If the 1.6 ratio falls toward ~1.1 and the Table II branching ratios move outside the quoted CLEO error bars, the '50% enhancement' and 'consistent with data' claims are not robust. Equivalently, quote the scale-variation band on Table II and on Eq. (16); if that band is comparable to the PQCD-versus-CLEO difference, the agreement is not meaningful.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's headline number is Eq. (16): (F^Pe6)_PQCD/(F^Pe6)_FA = 1.6, (F^Pe4)_PQCD/(F^Pe4)_FA = 1.4, (F_e)_PQCD/(F_e)_FA = 1.0. The text defines the FA baseline as 'the form factors evaluated in PQCD but with the Wilson coefficients C(t) set to C(M_B/2).' That makes the ratio, by construction, the integrand-weighted average of C6(t)/C6(M_B/2). Because C6 runs steeply and the Eq. (9) prescription t = max(√x M_B, 1/b1, 1/b3) pushes a large fraction of the integrand below m_b/2, the ratio is guaranteed to exceed unity. The physical content therefore lives entirely in whether that scale prescription is correct, not in the ratio itself. The paper justifies the prescription only by 'minimizes higher-order corrections,' citing [15], and quotes no scale-variation error. If the effective scale were closer to m_b/2, the ratio would move toward 1.0 and the '50%' would shrink. The abstract's phrasing 'compared to those assumed in the factorization approximation' also overstates the comparison: the baseline is a hybrid (PQCD integrals with frozen C), not full FA. A secondary tension: Table I gives F^P_a = 0.30 + 2.58i against F_e = 71.60, i.e. ~3.6% of the tree amplitude, so the physical amplitude's strong phase is only a few degrees; the claim of 'large CP asymmetries' is not quantified and is not obviously supported by these numbers.","agreement_with_reader":"agree"},"referee_report":{"model":"selfhosted/deepseek-v4.1-flash","summary":"The manuscript computes B → Kπ decay amplitudes in the perturbative QCD (PQCD) factorization framework of Li and collaborators, in which the amplitude is a convolution of B, K, and π light-cone wave functions with a hard kernel that includes Sudakov evolution. The authors give explicit integrands for the factorizable and annihilation amplitudes generated by the tree (O1,2), (V−A)(V−A) penguin (O3,4,9,10), and (V−A)(V+A) penguin (O5,6,7,8) operators (Eqs. 2–5), the hard functions (Eqs. 6–7), the scale-setting prescription (Eq. 9), and a specific set of meson wave functions (Eqs. 12–15). Three claims are made: (i) penguin amplitudes are dynamically enhanced by roughly 50% relative to a factorization-approximation baseline, because the PQCD scale choice pushes the integrand to t < m_b/2 where C4 and C6 are larger (Eq. 16); (ii) annihilation diagrams, in particular the O6 contribution F^P_a6, are not negligible and carry a strong phase near 90° (Table I); and (iii) the resulting B → Kπ branching ratios agree with CLEO data for a representative parameter set (Table II). The paper also argues that PQCD is self-consistent here because ~90% of the penguin amplitude comes from α_s(t)/π < 0.3, and contrasts its treatment with the BBNS approach.","tokens_in":11115,"tokens_out":5146,"duration_ms":92556,"significance":"If the central claims hold, the paper is a useful contribution: it offers a first-principles, explicitly reproducible calculation of B → Kπ amplitudes in which the penguin matrix elements and the annihilation strong phases are computed rather than assumed, and it makes falsifiable numerical predictions (Table II) that can be checked against the then-new CLEO data. The strengths worth naming are the fully explicit integrands and wave functions, which make the calculation reproducible; the perturbativity diagnostic in Fig. 3 and the accompanying fractional-contribution checks; and the authors' own acknowledgment that the strong phases carry roughly 20% uncertainty, translating into roughly 30% uncertainty in CP asymmetries. The comparison with BBNS is stated in a way that identifies a concrete, testable difference (the role of parton transverse momentum and Sudakov suppression in making annihilation diagrams leading-order).","major_comments":[{"comment":"The headline number of the paper, Eq. (16), is defined as a ratio of PQCD matrix elements computed with running Wilson coefficients C(t) to the same PQCD matrix elements computed with C(t) frozen at C(M_B/2). As the text states, '(F)_FA represent the form factors evaluated in PQCD but with the Wilson coefficients C(t) set to C(M_B/2).' The ratio is therefore, by construction, the integrand-weighted average of C6(t)/C6(M_B/2) and C4(t)/C4(M_B/2); it is guaranteed to exceed unity whenever the Eq. (9) prescription populates t < M_B/2, and it measures the scale choice rather than an independent dynamical effect. The physical content resides entirely in whether t = max(√x M_B, 1/b1, 1/b3) is the correct argument for the coefficients. The paper justifies this only by the statement that the choice 'minimizes higher-order corrections,' citing [15], and quotes no scale-variation band. A reader cannot tell from the manuscript whether the 1.6 and 1.4 factors are robust at the 10% level or are an artifact of one prescription. I recommend that the authors quote the ratio for at least one alternative scale choice (for example t = max(√x M_B, 1/b) with the 1/b terms dropped, or a fixed-scale variant), so that the reader can see the spread. Relatedly, the abstract's phrasing 'compared to those assumed in the factorization approximation' overstates the comparison: the baseline is a hybrid (PQCD integrals with frozen coefficients), not the full factorization approximation, and the text should say so explicitly in the abstract and in the sentence introducing Eq. (16).","section":"Eq. (16) and the definition of the FA baseline","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The abstract states that annihilation diagrams 'are not negligible, and give large strong phases,' and the text motivates this by the prospect of 'large CP asymmetries in the B → Kπ decays.' Table I, however, gives F^P_a = 0.30 + 2.58i against F_e = 71.60, i.e. the annihilation amplitude is about 3.6% of the factorizable tree amplitude. The phase of F^P_a is indeed near 90°, but the phase of the total amplitude is shifted by only a few degrees, and no CP asymmetry is computed anywhere in the paper. The claim as stated in the abstract is therefore not supported by the numbers the paper presents: 'large strong phase in F^P_a6' and 'large strong phase in the physical B → Kπ amplitude' are different statements, and only the first is demonstrated. I recommend that the authors either (a) present at least one explicit CP asymmetry (for example the direct asymmetry in B0 → K+π− versus B̄0 → K−π+) computed from the amplitudes in Table I, so that the reader can judge whether the annihilation contribution matters at an observable level, or (b) soften the abstract and the concluding paragraph to say that the annihilation amplitude carries a large phase but is a few-percent correction to the total amplitude at this order. The same paragraph should state explicitly which interference (tree–penguin, or penguin–annihilation) is responsible for the asymmetry being claimed.","section":"Table I and the annihilation claim (abstract s2/s6)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Table II quotes eight branching ratios with no theoretical uncertainty at all, alongside CLEO values with statistical and systematic errors, and the abstract concludes that the results are 'consistent with data.' Several entries are consistent within errors, but Br(B0 → K+π−) = 24.19 × 10^-6 against the quoted CLEO average 17.2^{+2.5}_{-2.4} ± 1.2 × 10^-6 is roughly a 2.5σ discrepancy, and Br(B+ → K0π+) = 21.72 against 18.2^{+4.6}_{-4.0} ± 1.6 is high by about 1σ. The blanket statement of consistency is therefore too strong for at least one mode. More importantly, the absence of any theory error bar makes the comparison uninterpretable: the inputs ω_B = 0.3 GeV, the Gegenbauer coefficients 0.8, 0.51, and 0.3 in Eqs. (13)–(14), m0π = 1.4 GeV, m0K = 1.7 GeV, and f_B = 190 MeV are all fixed numbers with no quoted range, and the scale ambiguity discussed above propagates directly into the penguin-dominated modes. I recommend that the authors attach an error estimate to Table II derived from varying ω_B, m0π, m0K, and the renormalization scale over plausible ranges, and that they qualify the consistency statement mode by mode. It would also help to state clearly which of the inputs in Eqs. (12)–(15) were determined from B → Dπ, ππ, and the pion form factor in [12] and which are predictions of this paper, since the Kπ modes are being compared to data using a parameter set that was itself tuned on other B decay data.","section":"Table II and the 'consistent with data' claim (abstract s7)","authors_acknowledge":null,"authors_acknowledge_p":null}],"minor_comments":[{"comment":"The second term in the square brackets of Eq. (7) is written '(b1 ↔ b3)', but the preceding expression involves b2 and b3, so the exchange should presumably read '(b2 ↔ b3)'. Please correct.","section":"Eq. (7)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"F^P_e is listed with a vanishing imaginary part. Please state explicitly whether absorptive parts from the charm-quark loop in the penguin matrix elements are included at this order, and if not, say so, since the reader may otherwise read the zero as a physical statement about the penguin strong phase.","section":"Table I","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The factor (3/2) e_q in the definition of a4 and a6 should be accompanied by a statement of which quark charges are summed over and with what weights, since the numerical results depend on it.","section":"Eq. (10)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The text states that annihilation and factorizable diagrams are 'both O(1/(M_B Λ_QCD))' and then that annihilation is 'subleading in the PQCD formalism as M_B → ∞' because the B-meson wave function enhances small x1. These two statements are in tension as written; please make the M_B-scaling argument explicit, since it is the crux of the disagreement with BBNS.","section":"Paragraph contrasting with BBNS","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Reference [12] is listed as 'to appear in Phys. Rev. D' and carries the determination of the wave functions and the φ3 ∼ 90° argument on which the penguin-enhancement motivation rests. Since the present manuscript depends on it for inputs, please confirm the published version at proof stage, or reproduce the few numbers needed to make the present letter self-contained.","section":"Reference [12]","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The fractional-contribution checks are quoted for F^P_e6 (90% and 80% below α_s/π = 0.3 and 0.2) and for F^Bπ (97% below 0.3), but not for F^P_a6 or F_e. Given that the average hard scale quoted in the text is about 1.4 GeV, please state the corresponding fractions for the annihilation amplitude as well, since that is the amplitude whose phase is being claimed as a prediction.","section":"Fig. 3 and the perturbativity discussion","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The intrinsic b dependence is retained for φ_B but dropped for φ_π and φ_K, with the justification that it produces only a 5% reduction in F^Bπ. Please state whether the same 5% estimate applies to the Kπ amplitudes, where the relevant scales are lower.","section":"Wave functions, Eqs. (12)–(15)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"There are several typographical errors that should be corrected: 'Comparision' in the section heading, 'outging' and 'meosn' in the FSI paragraph, 'calculaiton' in the BBNS comparison, and 'funciton' in the same paragraph. The figure axis labels in the supplied Fig. 3 are not legible and should be regenerated.","section":"General presentation","authors_acknowledge":null,"authors_acknowledge_p":null}],"recommendation":"major_revision","confidential_remarks":"The paper is a Letter-length announcement of a program whose details live in the companion paper [12], which is listed as 'to appear.' The calculation is explicit enough to be checked, and the perturbativity diagnostic is a genuine strength. The two issues that need work before the headline claims can stand as stated are the scale-prescription dependence of Eq. (16) and the gap between 'large phase in F^P_a6' and 'large CP asymmetry,' the latter never being computed. Neither is fatal, but both are load-bearing for the abstract."},"author_rebuttal":null,"desk_editor":{"model":"selfhosted/deepseek-v4.1-flash","letter":"The thing to know: this is the short letter that put numbers on the PQCD factorization program for B→Kπ, and the numbers are the point — penguin amplitudes 1.4–1.6× the frozen-scale baseline, an annihilation penguin with a strong phase near 90°, and an eight-channel branching-ratio table that lands on the CLEO values. The formalism is Li's, from the group's earlier papers; what is new here is the application and the specific results, and those results are not in the earlier work the paper cites. The paper is also candid about its own edges: the parameter set is called representative, the full scan is explicitly deferred, and the strong phases are quoted with a 20% uncertainty propagating to 30% in CP asymmetries. That is more honesty than most letters of this length bother with. The integrands are all written out (Eqs. 2–5), the wave functions and scale choices are stated, and the perturbativity check is not hand-waved — 90% of F_Pe6 comes from αs/π < 0.3, with the same shown for the form factor in Fig. 3. Where it is soft: the stress-test is right that Eq. (16) is a ratio of the same PQCD integral with running versus frozen Wilson coefficients. The body says so plainly — the FA baseline is defined as PQCD matrix elements with C(t) set to C(MB/2) — so the ratio is essentially the integrand-weighted average of C6(t)/C6(MB/2), and it exceeds unity by construction once Eq. (9) pushes weight below mb/2. The physics lives in whether that scale prescription is right, and the paper justifies it with one citation to a minimization argument and quotes no scale-variation band. The abstract's phrasing is looser than the body on this point. Table II also carries no theory errors at all. One place I would push back on the stress-test: the claim that the strong phase is only a few degrees compares F_Pa to F_e. CP asymmetries care about the tree-penguin relative phase, and with F_Pe = −6.18 and F_Pa = 0.30 + 2.58i the penguin amplitude carries roughly 20°, not a few. The 'large CP asymmetries' line is unquantified, but it is not obviously wrong. Who should read it: anyone working on B→Kπ or on the PQCD-versus-BBNS dispute. It is a letter, so the details live in the companion [12], cited as 'to appear' — worth having on hand when you check the integrals. Recommendation: send it out. The calculation is explicit, the prediction is falsifiable, and the disagreements are substantive rather than presentational.","headline":"An explicit, checkable PQCD calculation of B→Kπ whose headline 50% penguin enhancement is really a statement about the Eq. (9) scale choice, not an independent dynamical measurement.","tokens_in":11819,"tokens_out":2193,"would_cite":true,"duration_ms":44455,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"pith_extraction":{"msc":[],"pacs":["13.25.Hw","11.10.Hi","12.38.Bx","13.25.Ft"],"model":"selfhosted/deepseek-v4.1-flash","headline":"The paper argues that penguin amplitudes in $B$ meson decays are about 50% larger than the standard factorization approximation assumes, because the strong-interaction scale inside the decay is lower than usually chosen.","keywords":["B meson nonleptonic decays","perturbative QCD factorization","penguin operators","annihilation diagrams","strong phases","Wilson coefficients","CP violation","B to K pi branching ratios"],"falsifier":"Repeat the same PQCD integrals with the scale varied over a plausible range around $m_b/2$ and report the spread in $(F^{Pe6})_{PQCD}/(F^{Pe6})_{FA}$; if that ratio moves from about 1.6 down toward 1.0 within the allowed range, the central claim does not survive.","tokens_in":12041,"feed_emoji":"🐧","tokens_out":2820,"duration_ms":57817,"temperature":0.7,"pith_summary":"The paper asks how well the standard factorization approximation — which writes a two-body $B$ decay as a form factor times a decay constant — actually captures the underlying QCD. The authors use the perturbative QCD factorization framework, in which the decay amplitude is a convolution of meson wave functions with a hard scattering kernel evaluated at a scale set by the virtuality of the internal particles, and they apply it to $B^{0} to K^{-}pi^{+}$ and its relatives. Their main finding is that the penguin contributions, which are essential for explaining the observed $B to Kpi$ and $pipi$ rates, come out roughly 1.6 times larger (for the $O_6$ operator) and 1.4 times larger (for $O_4$) than the same calculation with the Wilson coefficients frozen at $M_B/2$, while the tree amplitude is unchanged. They also find that annihilation diagrams, conventionally dismissed as helicity-suppressed, are comparable in size to factorizable ones at $M_B sim 5$ GeV and carry a strong phase close to $90^circ$. The payoff claimed is a set of $B to Kpi$ branching ratios consistent with the CLEO measurements for a representative set of wave functions and quark masses.","feed_headline":"Penguins in B decays come out 50% fatter than assumed","feed_subtitle":"Choosing the interaction scale from internal-particle virtuality boosts penguin amplitudes and yields large strong phases.","key_machinery":"The carrying mechanism is perturbative QCD factorization for exclusive heavy-meson decays: the amplitude is written as a convolution of $B$, $K$ and $pi$ wave functions with a hard kernel $H([x],[b],M_B)$, with double logarithms resummed into Sudakov factors $exp[-S_B - S_pi]$ and $exp[-S_K - S_pi]$ that suppress the end-point region and keep the exchanged gluon hard. The scale-setting rule $t = max(sqrt{x},M_B, 1/b_i, 1/b_j)$ is what converts the running of $C_4$ and $C_6$ into the numerical enhancement, and the same framework makes annihilation diagrams formally the same order in $1/M_B$ as factorizable ones, which is what allows them to generate a large imaginary part.","core_discovery":"The central claim is a dynamical enhancement of penguin amplitudes in $B to Kpi$ decays. The mechanism is the running of the Wilson coefficients $C_4$ and $C_6$, which depend steeply on the renormalization scale, combined with the prescription that the hard scale be taken as the maximum of $sqrt{x},M_B$ and the inverse transverse extents $1/b$ of the participating mesons. Because a large fraction of the integrand sits at $t < m_b/2$, the penguin coefficients are evaluated where they are larger, giving $(F^{Pe6})_{PQCD}/(F^{Pe6})_{FA} = 1.6$, $(F^{Pe4})_{PQCD}/(F^{Pe4})_{FA} = 1.4$, and $(F^{e})_{PQCD}/(F^{e})_{FA} = 1.0$. A second claim is that the penguin annihilation amplitude $F^{Pa6}$ is not negligible: in Table I it has real part $0.30$ and imaginary part $2.58$ in units of $10^{-2}$, i.e. a strong phase near $90^circ$, arising from cuts on the intermediate $sbar{d}$ state. The authors argue this is allowed because the $O_6$ operator's helicity structure lets a left-handed quark and left-handed antiquark emerge back to back, so the usual helicity-suppression argument for annihilation does not apply. They check perturbativity by showing that about 97% of the $B to pi$ form factor and roughly 90% (80%) of $F^{Pe6}$ come from the region $alpha_s(t)/pi < 0.3$ (0.2).","pith_inferences":["If the enhancement is really a scale-choice effect rather than a dynamical one, then the same prescription applied to other channels with steeply running coefficients — charmless $B to pipi$, $B to Keta$, $B to K K$ — should show a correlated pattern of shifts, which is a sharper test than any single branching ratio.","The near-$90^circ$ annihilation phase suggests measurable direct CP asymmetries in $B^pm to K^pm pi^0$ and $B^0 to K^pm pi^mp$ that are larger than factorization-based expectations; comparing the two charge-conjugate rates is the natural place to look.","Because the enhancement rests on integrand weight below $m_b/2$, an independent determination of the effective scale from, say, the $B to pi$ form factor's shape at large recoil would either corroborate or undercut the claim without needing new decay data.","The paper's perturbativity check is a statement about the fraction of the integrand in a small-$alpha_s$ region, not a bound on the neglected higher-order terms; a genuine next-to-leading calculation would be the way to convert that fraction into an error bar."],"forward_implications":["The $B to Kpi$ and $B to pipi$ branching ratios can be fitted with a unitarity angle $phi_3 sim 90^circ$ using one consistent set of parameters, which the authors say is difficult in the plain factorization approximation.","Penguin matrix elements extracted from data under the factorization approximation are systematically underestimated by roughly a third to a half.","Annihilation diagrams must be retained in any CP-asymmetry prediction for $B to Kpi$, since they supply strong phases of order $90^circ$ rather than negligible ones.","The strong phases are computable in this framework to roughly 20%, which the authors translate into about 30% uncertainty on predicted CP asymmetries.","The relative weight of annihilation versus factorizable contributions should shrink as $M_B to infty$, giving a concrete asymptotic check on the picture."],"supporting_citations":[{"why":"CLEO branching-ratio measurements that the predicted $B to Kpi$ rates are compared against in Table II.","marker":"[3]"},{"why":"Supplies the perturbative QCD factorization formalism and Sudakov exponents used for exclusive $B$ decays.","marker":"[4]"},{"why":"Original Brodsky-Lepage hard-scattering formalism on which the convolution of wave functions with a hard kernel rests.","marker":"[8]"},{"why":"Botts-Sterman resummation of transverse-momentum logarithms that provides the end-point suppression making the calculation self-consistent.","marker":"[9]"},{"why":"Source of the meson wave functions, decay constants, and $m_0$ parameters used for the numerical results in Tables I and II.","marker":"[12]"},{"why":"Justifies the choice of hard scale as the internal-particle virtuality that minimizes higher-order corrections.","marker":"[15]"},{"why":"QCD sum-rule determination of the kaon wave function employed in the numerical evaluation.","marker":"[17]"},{"why":"The competing factorization-based approach the paper explicitly contrasts itself with on the treatment of annihilation and form factors.","marker":"[18]"},{"why":"Provides the detailed argument for why annihilation diagrams are not suppressed and can generate large imaginary parts.","marker":"[21]"}],"fun_headline_variants":["Penguin amplitudes in B to K pi are 50% larger than factorization assumes","Scale choice makes penguin terms in B to K pi about 50% larger","Annihilation diagrams are not negligible in B to K pi decays","Annihilation penguin in B to K pi has a strong phase near 90 degrees","Representative parameter set brings B to K pi rates in line with data"],"cache_read_input_tokens":0,"weakest_assumption_plain":"Everything hinges on taking the hard scale $t$ as the maximum of the internal-particle virtuality and the inverse transverse sizes, since the quoted 50% enhancement is literally the difference between evaluating the Wilson coefficients at that scale and freezing them at $M_B/2$.","fun_headline_variants_meta":{"raw":{"variants":["Penguin amplitudes in B to K pi are 50% larger than factorization assumes","Scale choice makes penguin terms in B to K pi about 50% larger","Annihilation diagrams are not negligible in B to K pi decays","Annihilation penguin in B to K pi has a strong phase near 90 degrees","Representative parameter set brings B to K pi rates in line with data"]},"model":"selfhosted/deepseek-v4.1-flash","effort":"low","attempts":1,"cost_usd":0.0,"raw_usage":{"total_tokens":2077,"prompt_tokens":1025,"completion_tokens":1052,"prompt_tokens_details":null,"completion_tokens_details":{"reasoning_tokens":964}},"tokens_in":1025,"tokens_out":1052,"duration_ms":17458,"image_pages":[],"temperature":1.0,"reasoning_budget":50,"reasoning_tokens":964,"transport_provider":"selfhosted","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-10-29T18:13:41.557585+00:00","model_set":{"reader":"selfhosted/deepseek-v4.1-flash"},"falsifier":"Repeat the same PQCD integrals with the scale varied over a plausible range around $m_b/2$ and report the spread in $(F^{Pe6})_{PQCD}/(F^{Pe6})_{FA}$; if that ratio moves from about 1.6 down toward 1.0 within the allowed range, the central claim does not survive.","supporting_citations":[{"cited_title":"Study of Two-Body B Decays to Kaons and Pions: Observation of B->pi+pi-, B-> K+-pi0, and B-> K0pi0 Decays","cited_arxiv_id":"hep-ex/0001010","evidence_quote":"CLEO branching-ratio measurements that the predicted $B to Kpi$ rates are compared against in Table II."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the perturbative QCD factorization formalism and Sudakov exponents used for exclusive $B$ decays."},{"cited_title":"Phys. Rev. D 22, 2157","cited_arxiv_id":null,"evidence_quote":"Original Brodsky-Lepage hard-scattering formalism on which the convolution of wave functions with a hard kernel rests."},{"cited_title":"Nucl. Phys. B225, 62","cited_arxiv_id":null,"evidence_quote":"Botts-Sterman resummation of transverse-momentum logarithms that provides the end-point suppression making the calculation self-consistent."},{"cited_title":"Penguin Enhancement and $B\\to K\\pi$ decays in perturbative QCD","cited_arxiv_id":"hep-ph/0004173","evidence_quote":"Source of the meson wave functions, decay constants, and $m_0$ parameters used for the numerical results in Tables I and II."},{"cited_title":"Phys. Rev. D 60, 074004","cited_arxiv_id":null,"evidence_quote":"Justifies the choice of hard scale as the internal-particle virtuality that minimizes higher-order corrections."},{"cited_title":"JHEP 9809, 005","cited_arxiv_id":null,"evidence_quote":"QCD sum-rule determination of the kaon wave function employed in the numerical evaluation."},{"cited_title":"QCD factorization for exclusive, non-leptonic B meson decays: General arguments and the case of heavy-light final states","cited_arxiv_id":"hep-ph/0006124","evidence_quote":"The competing factorization-based approach the paper explicitly contrasts itself with on the treatment of annihilation and form factors."},{"cited_title":"Nonleptonic charmless B decays: factorization vs perturbative QCD","cited_arxiv_id":"hep-ph/0006001","evidence_quote":"Provides the detailed argument for why annihilation diagrams are not suppressed and can generate large imaginary parts."}],"review_version":1}